PDF table to Excel
Extract tables from PDF to Excel
"Convert PDF to Excel" usually means "give me the whole page as cells and good luck". If you only wanted the table, that's a lot of deleting. With doc2json you name the columns you care about, and you get just those rows back as a clean Excel sheet. Price lists, rate cards, schedules, parts lists, anything that's a table trapped in a PDF.
Made-up example
What the Excel file looks like
Example only. Northwind Supply's price list is invented.
{
"type": "object",
"properties": {
"valid_from": {
"type": "date",
"description": "Date the prices apply from"
},
"rows": {
"type": "array",
"items": {
"type": "object",
"properties": {
"item_code": {
"type": "string"
},
"description": {
"type": "string"
},
"unit": {
"type": "string"
},
"price": {
"type": "number",
"description": "Unit price, without currency symbol"
}
}
}
}
}
}| Row | document | valid_from |
|---|---|---|
| 2 | northwind-prices-nov.pdf | 2026-11-01 |
Four columns asked for, four columns delivered. The logo, the address block and "Prices subject to change" stayed in the PDF where they belong.
How it works
Three steps, no setup.
- Step 1
Describe the table
In the schema editor, add a list field (say rows) and the columns inside it, like item_code, description and price. A short description per column helps with odd headers.
- Step 2
Upload the PDF
Digital PDF, scan or photo. doc2json reads every page and fills your rows in order.
- Step 3
Download Excel
Your rows land on their own sheet, numbers as numbers, dates as dates. No merged cells, no stray headers.
API
One request, one workbook
Add export=xlsx to the parse. The JSON comes back as usual, with the Excel file next to it as base64, for the same credits.
No saved schema? Pass the schema JSON itself as the schema field. Want CSV? Use export=csv and you get one row per table row, with valid_from repeated on each.
Read the API docscurl -X POST https://doc2json.manoonlabs.com/api/v1/parse \
-H "Authorization: Bearer $DOC2JSON_KEY" \
-F file=@northwind-prices-nov.pdf \
-F schema=price-list \
-F export=xlsx \
| jq -r '.export.content' | base64 -d > northwind-prices-nov-2026-10-30.xlsxWhy doc2json
Why doc2json and not a generic PDF to Excel converter
You pick the columns
Generic converters export everything on the page. You list the columns once and get only those, in the order you defined.
One table, even across pages
A table that continues over several pages comes back as one list of rows, with repeated header rows dropped.
No merged-cell mess
Wrapped text stays in one cell, every row has the same columns, and numbers are numbers you can sort and sum.
FAQ
PDF table to Excel, answered.
Can I extract a table from a scanned PDF?
Yes. Scans and photos are read the same way as digital PDFs, as long as the text is legible.
The PDF has three tables. Can I get all of them?
Yes. Add one list field per table to your schema (for example prices and discounts) and each one gets its own sheet in the Excel file.
Do I have to set up anything per document layout?
No. There's no drawing boxes around tables. The same schema works across PDFs with different layouts, because doc2json reads the document rather than matching positions.
Is my PDF stored?
No. The PDF is processed in memory for the request and then dropped. Neither the file nor the extracted data is saved, only a usage record with the time, page count and credits used.
What does PDF table extraction cost?
The free plan gives you 50 pages a month, one credit per page. Exporting to Excel or CSV is included at no extra cost.
Related
- Working with invoices? A ready-made schema is on Invoice to Excel
- Bank statements? Bank statement to Excel
- All templates
Free your table from the PDF.
Describe the columns once and reuse them forever. 50 free pages a month.